Re: [PATCH v3] pipe: use __pipe_{lock,unlock} instead of spinlock
From: Sedat Dilek
Date: Fri Jan 06 2023 - 15:35:11 EST
On Fri, Jan 6, 2023 at 8:40 PM Luis Chamberlain <mcgrof@xxxxxxxxxx> wrote:
>
> On Fri, Jan 06, 2023 at 05:48:44PM +0800, Hongchen Zhang wrote:
> > Use spinlock in pipe_read/write cost too much time,IMO
> > pipe->{head,tail} can be protected by __pipe_{lock,unlock}.
> > On the other hand, we can use __pipe_{lock,unlock} to protect
> > the pipe->{head,tail} in pipe_resize_ring and
> > post_one_notification.
> >
> > I tested this patch using UnixBench's pipe test case on a x86_64
> > machine,and get the following data:
> > 1) before this patch
> > System Benchmarks Partial Index BASELINE RESULT INDEX
> > Pipe Throughput 12440.0 493023.3 396.3
> > ========
> > System Benchmarks Index Score (Partial Only) 396.3
> >
> > 2) after this patch
> > System Benchmarks Partial Index BASELINE RESULT INDEX
> > Pipe Throughput 12440.0 507551.4 408.0
> > ========
> > System Benchmarks Index Score (Partial Only) 408.0
> >
> > so we get ~3% speedup.
> >
> > Signed-off-by: Hongchen Zhang <zhanghongchen@xxxxxxxxxxx>
> > ---
>
> After the above "---" line you should have the changlog descrption.
> For instance:
>
> v3:
> - fixes bleh blah blah
> v2:
> - fixes 0-day report by ... etc..
> - fixes spelling or whatever
>
> I cannot decipher what you did here differently, not do I want to go
> looking and diff'ing. So you are making the life of reviewer harder.
>
Happy new 2023.
Positive wording... You can make reviewers' life easy when...
(encourage people).
Life is easy, people live hard.
+1 Adding ChangeLog of patch history
Cannot say...
Might be good to add the link to Linus test-case + your results in the
commit message as well?
...
Link: https://git.kernel.org/linus/0ddad21d3e99 (test-case of Linus
suggested-by Andrew)
...
Signed-off-by: Hongchen Zhang <zhanghongchen@xxxxxxxxxxx>
...
Thanks.
Best regards,
-Sedat-